Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/cryptopilot-tools/cryptopilot-mcp/claude-mdgit clone --depth 1 https://github.com/cryptopilot-tools/cryptopilot-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/cryptopilot-tools/cryptopilot-mcp/claude-md)<a href="https://agentmods.dev/instructions/cryptopilot-tools/cryptopilot-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/cryptopilot-tools/cryptopilot-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01640 | $0.01640 |
| Opus 5 | $0.00820 | $0.00820 |
| Sonnet 5 | $0.00328 | $0.00328 |
| Haiku 4.5 | $0.00164 | $0.00164 |
Grade A, and why
cryptopilot-mcp CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 207 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Context file for AI coding agents working on this repository. Read this first when starting a session.
Project
CryptoPilot MCP — A read-only Model Context Protocol (MCP) server that exposes crypto and brokerage portfolio data to Claude Desktop and other MCP-compatible clients.
Project boundary: CryptoPilot is a completely independent project. Agents working in this repository must not reference, compare, or assume any relationship with other projects (including stockpilot-mcp). Each project has its own scope, codebase, and decisions.
Stack
- Language: TypeScript (strict mode)
- Runtime: Node.js 18+
- MCP SDK:
@modelcontextprotocol/sdkv1.29+ - Transport: stdio
- Validation: Zod
- Auth methods: JWT ES256 (Coinbase), HMAC SHA256 (SnapTrade)
- Package manager: npm
Architecture
Claude Desktop communicates with the MCP server over stdio. The server
uses an abstract BaseProvider class, with concrete implementations for
Coinbase (direct Advanced Trade API) and SnapTrade (which aggregates
Robinhood and 30+ other brokerages).
Each provider has its own normalizer that maps raw API responses to
canonical types (CanonicalAccount, CanonicalHolding).
Key design decisions:
costBasis(total invested) andaveragePurchasePrice(per unit) are separate fields inCanonicalHoldingto handle high-priced stocks like BRK.A correctly.- HTTP retry with exponential backoff (1s, 2s, 4s) for 429/5xx responses.
- Coinbase
balance.amountis always 0 in their API; actual balance isavailable + hold.
MCP Tools (5)
list_providers— returns configured providers and capabilitieslist_accounts— returns accounts (optionalproviderfilter)list_holdings— returns holdings for an account (requiresaccountId)get_quote— returns current price (requiresprovider+symbol)get_provider_health— health check for each configured provider
Out of scope (Phase 1): transactions, trading, withdrawals. This is read-only by design.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 207 lines · 1,640 tokens per session scan A e309245ace99
cryptopilot-mcp CLAUDE.md is an instructions file published in the GitHub repository cryptopilot-tools/cryptopilot-mcp (0 stars, last pushed 2mo ago), licensed MIT. It adds 1,640 tokens to every session, about $0.0082 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
gmgn-skills CLAUDE.md
Instructions for GMGNAI/gmgn-skills, covering claude.md, critical rule — read this first, project overview, available skills and quick decision guide.
coinos-skills CLAUDE.md
Instructions for aicoincom/coinos-skills, covering claude.md, project overview, architecture, coinclaw 三引擎适配 and 关键设计决定.
whitebit-net
Use WhiteBit.Net when generating C#/.NET code that interacts with the WhiteBit cryptocurrency exchange, including V4 REST endpoints, WebSocket subscriptions, spot trading, collateral/futures trading, balances, deposits, withdrawals, convert, WhiteBit Codes, sub-accounts, and CryptoExchange.Net shared APIs.
defi-agents copilot-instructions.md
Copilot instructions for nirholas/defi-agents, covering defi-agents and terminal management.
Binance-US-MCP CLAUDE.md
Claude Code instructions for nirholas/Binance-US-MCP, covering binance-us-mcp and terminal management.
tiingo-mcp AGENTS.md
Instructions for major7apps/tiingo-mcp, covering agents.md, project map, public contract, operating rules and commands.